Limited Sampling Strategies to Predict the Area Under the Concentration-Time Curve for Rifampicin

Background: Rifampicin (RMP) is the most effective first-line antituberculosis drug. One of the most critical aspects of using it in fixed-drug combination formulations is to ensure it reaches therapeutic levels in blood. The determination of the area under the concentration-time curve (AUC) and appropriate dose adjustment of this drug may contribute to optimization of therapy. Even when the maximal concentration (Cmax) of RMP also predicts its sterilizing effect, the time to reach it (Tmax) takes 40 minutes to 6 hours. The aim of this study was to develop a limited sampling strategy (LSS) for therapeutic drug monitoring assistance for RMP. Methods: Full concentration-time curves were obtained from 58 patients with tuberculosis (TB) after the oral administration of RMP in fixed-drug combination formulation. A validated high-performance liquid chromatographic method was used. Pharmacokinetic parameters were estimated with a noncompartmental model. Generalized linear models were obtained by forward steps, and boot-strapping was performed to develop LSS to predict AUC curve from time 0 to the last measured at 24 hours postdose (AUC(0-24)). The predictive performance of the proposed models was assessed using RMP profiles from 25 other TB patients by comparing predicted and observed AUC(0-24). Results: The mean AUC(0-24) in the current study was 91.46 +/- 36.7 mg.h.L-1, and the most convenient sampling time points to predict it were 2, 4 and 12 hours postdose (slope [m] = 0.955 +/- 0.06; r(2) = 0.92). The mean prediction error was -0.355%, and the root mean square error was 5.6% in the validation group. Alternate LSSs are proposed with 2 of these sampling time points, which also provide good predictions when the 3 most convenient are not feasible. Conclusions: The AUC(0-24) for RMP in TB patients can be predicted with acceptable precision through a 2- or 3-point sampling strategy, despite wide interindividual variability. These LSSs could be applied in clinical practice to optimize anti-TB therapy based on therapeutic drug monitoring.